Path: news.net.uni-c.dk!howland.erols.net!dispose.news.demon.net!demon!colt.net!newsfeed.icl.net!nntp.news.xara.net!xara.net!gxn.net!cygnus.co.uk!not-for-mail
From: aph@redhat.invalid
Newsgroups: comp.ai.neural-nets,comp.lang.apl,comp.lang.awk,comp.lang.beta,comp.lang.cobol,comp.lang.dylan,comp.lang.forth
Subject: Re: Einstein's Riddle
Date: Wed, 14 Mar 2001 16:53:50 +0000 (UTC)
Organization: Red Hat UK
Lines: 45
Message-ID: <98o7mu$vcn$1@localhost.localdomain>
References: <VIPq6.396633$ge4.138469576@news2.rdc2.tx.home.com> <3AACB567.A59B8497@Azonic.co.nz> <3AACE6CF.7F05484D@ieee.org> <0W8r6.178$fo5.14165@news.get2net.dk> <3AAD60F3.120F284A@ieee.org> <3AAE371A.2F9F596F@brazee.net> <98m43a$fe2$1@localhost.localdomain> <3AAEAD1A.BCDE11DB@ix.netcom.com> <98mugg$2mj$1@news.igs.net> <3AAF13CA.C7EA3113@ix.netcom.com> <98njv5$rfh$1@localhost.localdomain> <3AAF9AB6.2E06BEBD@ix.netcom.com>
NNTP-Posting-Host: viagra.cambridge.redhat.com
X-Trace: localhost.localdomain 984588830 32151 172.16.18.77 (14 Mar 2001 16:53:50 GMT)
X-Complaints-To: usenet@localhost.localdomain
NNTP-Posting-Date: Wed, 14 Mar 2001 16:53:50 +0000 (UTC)
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.16-22 (i686))
Xref: news.net.uni-c.dk comp.ai.neural-nets:67585 comp.lang.apl:29405 comp.lang.awk:17172 comp.lang.beta:12770 comp.lang.cobol:102665 comp.lang.dylan:24191 comp.lang.forth:78590

In comp.lang.forth J Thomas <jethomas@ix.netcom.com> wrote:
: aph@redhat.invalid wrote:

:> Perhaps, but it's no use as a test: a computer can easily be 
:> programmed to make mistakes, and a computer programmed using 
:> heuristics will definitely make "mistakes".

: It may not be easy to program it to make the kind of mistakes that are
: recognisably human.  But apart from that, how is it an intelligence test
: if you have to go to extra lengths to make your program make mistakes?

I don't think that's necessary.  An AI sufficiently "advanced" to pass
a Turing Test might well make lots of mistakes too.

:> Turing's point is simply that it is absurd to credit a person with
:> intelligence if you refuse, given equal evidence, to credit a 
:> computer.

: OK!  Rhat's a good philosophical point.  Turing tells us that we don't
: have a clear concept of intelligence,

Yes; the Turing Test is an attempt to refine our concept of
intelligence.

: that defining intelligence as "what humans do" is silly.

I don't think so.  Or, how do you make this out?

:> : I doubt that there's a unitary intelligence.  Different brains are 
:> : good at solving different problems.  We won't be ready to 
:> : understand the intelligence of oak trees

:> There is no evidence that oak trees do anything related to cognitive
:> processing.

: We haven't found any evidence.

Right.  

: We *won't* find any evidence unless we get a feel for what problems
: they solve.

So we'll probably never know.  Never mind.

Andrew.